RJ10349 (A0506-007) June 20, 2005 Other IBM Research Report Levels of Business Structures Representation K. Grigorova, P. Hristova, G. Atanasova Department of Informatics and Information Technologies University of Rousse 8 Studentska Str. Rousse, Bulgaria J. Q. Trelewicz IBM Research Division Almaden Research Center 650 Harry Road San Jose, CA 95120-6099 Research Division Almaden - Austin - Beijing - Haifa - India - T. J. Watson - Tokyo - Zurich LIMITED DISTRIBUTION NOTICE: This report has been submitted for publication outside of IBM and will probably be copyrighted if accepted for publication. I thas been issued as a Research Report for early dissemination of its contents. In view of the transfer of copyright to the outside publisher, its distribution outside of IBM prior to publication should be limited to peer communications and specific requests. After outside publication, requests should be filled only by reprints or legally obtained copies of the article (e.g ,. payment of royalties). Copies may be requested from IBM T. J. Watson Research Center , P. O. Box 218, Yorktown Heights, NY 10598 USA (email: [email protected]). Some reports are available on the internet at http://domino.watson.ibm.com/library/CyberDig.nsf/home . 1 LEVELS OF BUSINESS STRUCTURES REPRESENTATION K. Grigorova, P. Hristova, G. Atanasova. J. Q. Trelewicz Abstract: With the increasingly large number of software frameworks available to facilitate "business modeling", it is important to understand the implications of the level of abstraction provided by the frameworks. In this paper, we discuss three levels of abstraction of framework for business modeling, including the most typical and widely accepted representatives of each level. We show that XML is an emerging standard of data interchange and integration for business modeling. We discuss these frameworks in the context of database representation, which is important for storage and retrieval of models. Keywords: Business Modeling, Modeling Languages, Modeling Techniques. Introduction Competitive pressure, globalization, and the wide availability of Internet have made necessary the formal design of businesses. While in the past, business practices -- rules, routines, procedures, and processes -- could evolve in a piecemeal, isolated, and historical way, today, a rigorous and systemic design of such practices is needed, to ensure that customers' requests for products and services are processed at the satisfactory speed. Today’s business modeling aims at the integration of the partial models that represent particular views on an enterprise. This means not only that models of distinctive and important parts of the enterprise should be created, but also that semantic relationships between partial models can be expressed. The basic idea of business modeling is to offer different views on the business. The views should complement each other and thereby support a better understanding of complex systems by emphasizing appropriate abstractions. Therefore a corresponding modeling language is used based on specific terminology that is common within particular view. It provides intuitive concepts to structure the problem domain in a meaningful way. The baseline views should be flexible in the sense that they can be applied to any business area. Then business modeling may provide concepts that can be reused and adapted in a convenient way to detailed models for businesses in a specific market. Specialized modeling languages are one example. Other examples include reference models for certain types of industry. Business modeling is performed on different levels of detail according to the needs of designers. Sometimes it is sufficient to create a common picture of the enterprise, while other times the use of detailed concept is required. For this reason the business modeling methods should allow various levels of abstraction: • The highest level in the hierarchy of abstraction is related to external description allowing the users to express their view of how a given business structure looks, a type of meta-modeling language. Often, communication between people that belong to different professional communities will not require a high level of detail. Instead it is sufficient, and helpful, to seek a common understanding of the "big picture". On the other hand, there are also specific tasks, like the re-design of a business process or the design of an object model that require the use of detailed concepts. This confirms the existence of variety of modeling languages and techniques. They are used to create the meta-model of a given business structure. • The next level in the hierarchy of abstraction considers internal representation of business structures, which is a kind of application level. This level of representation corresponds to the requirements of a given application. The variety of modeling languages and techniques does not demand the multiplicity of internal descriptions. It is preferable to use some standard approaches in order to allow the successful exchange of models between different environments. Lately the emergence of XML (eXtensible Markup Language) is a first step to solve the problem of the variety modeling languages. XML is now widely accepted and acknowledged a standard. XML allows representing information in a simple, readable format, easily parsed by software tools. • The third, and final, level of the hierarchy of abstraction is related to the real (existing) storage. 2 The representation on this level may be viewed in some sense as on-line database. In some implementations, the user is not allowed to access it and its format may not even be known by the user. In other implementations, it might be useful to enable the system analysts to operate directly with database models. Additionally, the effective database design leads to successful performance. In this paper we discuss both external and internal levels of business structures representation and indicate the corresponding database models as a proper subject of efficient analysis. Sometimes we refer to modeling techniques suitable mostly for representation of business processes as the most important part of business modeling. External representation of business structures There is a large variety of meta-languages commonly used for the representation of the highest level of business structures. Generally these meta-languages involve different graphical primitives for describing the objects and connections between them. The most important aspect of all of the meta-languages is important part that the notation plays in any model – it is the glue that holds the process together. Notation has three roles: • It serves as the language for communicating decisions that are not obvious or cannot be inferred from the core itself; • It provides semantics that are rich enough to capture all important strategic and tactical decisions; • It offers a form concrete enough for humans to reason and for tools to manipulate. Here we will discuss some typical and wide spread representatives of modeling tools, used for external business structures description. Flowcharting Flowcharting is among the first graphical modeling techniques, dating back to the 1960s. The advantages of flowcharts centre on their ability to show the overall structure of a system, to trace the flow of information and work, to depict the physical media on which data are input, output and stored, and to highlight key processing and decision points [Schriber, 1969] [Jones, 1986]. Flowcharting was initially intended to provide computer program logic representation, but, because of its flexible nature, it has been used in many other application areas as well, including business modeling. Despite its advantages, namely familiarity and ease of use, flowcharting is no longer a dominant modeling technique because it can provide only basic facilities in representing processes. Therefore, in the area of business modeling, flowcharts nowadays are typically used primarily as a simple, graphic means of communication, intended to support narrative descriptions of processes, when the latter become complicated and difficult to follow. Data flow diagrams Data Flow Diagramming (DFD) is a technique for graphically depicting the flow of data amongst external entities, internal processing steps, and data storage elements in a business process. DFDs are used to document systems by focusing on the flow of data into, around, and outside the system boundaries. In that respect, DFDs are comparable to flowcharts, differing from them basically in the focus of analysis: DFDs focus on data, instead of activities and control [Yourdon, 1989]. DFDs have been widely used for data modeling purposes and have become an ad-hoc standard notation for traditional systems analysis and design. DFDs used to model the system’s data processing and the flow of information from one process to another. They are an intrinsic part of many analysis methods. They show the sequence of processing steps traversed by the data. Each step documents an action taken to transform or distribute the data. DFDs are easy to read, making it possible for domain experts to create or to validate the diagrams [Sommerville, 2003]. Entity-Relationship diagrams Entity-Relationship (ER) diagrams [Yourdon, 1989] are another widely used data modeling technique. ER diagrams are network models that describe the stored data layout of a system. ER diagrams focus on modeling the data present in a system and their inter-relationships in a manner that is entirely